@edupia-tutor/spec-driven-docs 0.14.0

package/skills/debug/SKILL.tmpl

@@ -0,0 +1,262 @@
+---
+description: Fixes bugs with full workflow (branch, test, commit), performs quick debug analysis of errors or unexpected behavior, or validates traceability coverage between specs and code. Trigger when: "/fix-bug", "/debug", "/validate-traces", "fix bug", "sửa bug", "debug lỗi", "phân tích lỗi", "tại sao lỗi này", "validate traces", "kiểm tra traceability", "coverage matrix", "trace drift".
+---
+
+# Debug & Quality Skills — Fix Bug, Debug, Validate Traces
+
+This skill handles: `/fix-bug` (full bug fix workflow), `/debug` (quick analysis), and `/validate-traces` (traceability coverage check).
+
+---
+
+## /fix-bug — Full Bug Fix Workflow
+
+### Input
+
+Accept: ticket ID (e.g. `PROJ-123`) or a direct bug description.
+
+### Phase 1 — Gather Information
+
+If ticket ID provided:
+- Fetch ticket details if Jira/issue tracker is connected
+- Otherwise ask user to paste: title, steps to reproduce, expected vs actual
+
+If no ticket:
+**CHECKPOINT** — Ask:
+1. Where does the bug occur? (service, endpoint, flow)
+2. Steps to reproduce?
+3. Expected vs Actual behavior?
+4. Error log / stack trace (if any)?
+
+### Phase 2 — Root Cause Analysis
+
+{{include:steps/context-loader.md}}
+
+Trace through layers (from CLAUDE.md architecture):
+1. Identify the service/module from the failing endpoint/feature
+2. Read relevant code files, layer by layer
+3. Match against common bug patterns:
+
+| Bug Type | Common Location | How to Check |
+|----------|----------------|--------------|
+| Wrong response data | Data mapping layer | Check field mapping, DTO conversions |
+| 400 Bad Request | Input validation | Check DTO constraints, required fields |
+| 403 Forbidden | Auth configuration | Check role-based access rules |
+| 404 Not Found | Repository query | Check find method, ID types |
+| Slow query / N+1 | Data access layer | Check for missing JOIN FETCH or batch loading |
+| Null Pointer | Missing null checks, optional not handled | Check Optional.orElseThrow usage |
+| Transaction rollback | Missing/wrong transaction scope | Check @Transactional on service methods |
+| Stale cache | Missing cache eviction after write | Check cache invalidation on write operations |
+| Type mismatch in query | Filter/specification code | Check field types in query predicates |
+| Cross-service timeout | External service call | Check URL config, fallback behavior |
+
+**CHECKPOINT — Root Cause Report:**
+
+```
+Root Cause Analysis — {TICKET_ID}:
+
+Bug: {description}
+Affected module: {module/service}
+Root cause: {specific technical analysis}
+Affected files:
+  - {file}: {what's wrong}
+
+Proposed fix:
+  - {file}: {what to change}
+
+Regression risk: Low / Medium / High
+  → Reason: {explanation}
+
+Proceed with fix? (Y/N)
+```
+
+Wait for user confirmation.
+
+### Phase 3 — Implement Fix
+
+```bash
+git checkout -b fix/{TICKET_ID}-{short-description}
+```
+
+Apply the fix exactly as analyzed. Follow CLAUDE.md coding standards — do not introduce new violations.
+
+Add traceability annotation if the file has `@trace.implements`:
+```
+// @trace.implements={UC-ID}-SC{N}
+// @trace.fixes={TICKET_ID}
+// @trace.root_cause={brief description}
+```
+
+### Phase 4 — Regression Test
+
+Generate a regression test if one doesn't exist:
+```
+// @trace.verifies={UC-ID}
+// @trace.regression={TICKET_ID}
+Test: "Regression {TICKET_ID}: {bug description}"
+  → Reproduce the exact bug scenario
+  → Assert the fix works correctly
+```
+
+Run the test: `{TEST_COMMAND_FOR_CLASS}`
+
+If test fails → debug and fix (max 3 iterations).
+
+### Phase 5 — Build Verify
+
+```bash
+{BUILD_COMMAND}   # max 3 retries
+```
+
+### Phase 6 — Commit & Push
+
+```bash
+git add {changed files}
+git commit -m "fix({TICKET_ID}): {brief description}"
+git push -u origin fix/{TICKET_ID}-{slug}
+```
+
+### Output
+
+```
+/fix-bug Complete — {TICKET_ID}
+
+## Root Cause
+{analysis}
+
+## Changes
+- {file}: {what changed and why}
+
+## Regression Test
+✅ Added: {TestClass}.{method}
+✅ Build: SUCCESS
+
+## Branch
+fix/{TICKET_ID}-{slug}
+```
+
+{{include:steps/report-footer.md}}
+
+---
+
+## /debug — Quick Debug Analysis
+
+Use when: IDE shows an error, test fails unexpectedly, behavior is strange, or you need to understand why code does something.
+
+**Different from `/fix-bug`**: This is analysis only, no full workflow, no ticket required.
+
+### Input
+
+Paste one of:
+1. Stack trace / error log
+2. Test failure output
+3. File path + description of unexpected behavior
+4. A specific question about a code snippet
+
+### Stack Trace Analysis
+
+Read stack trace from **bottom up** — the `Caused by:` line is the real root cause.
+
+```
+Caused by: {RealException}  ← start here
+  at {class}.{method}({file}:{line})  ← trace from here
+  ...
+  at {entrypoint}  ← where it was called from
+```
+
+### Common Error Patterns
+
+| Error | Likely Cause | Fix Direction |
+|-------|-------------|---------------|
+| NullPointerException | Accessing field on null object; Optional not handled | Check Optional.orElseThrow, null guards |
+| ClassCastException | Wrong type assumption; incorrect generic usage | Check type at assignment/return point |
+| OutOfMemoryError | Loading too much data; memory leak | Add pagination, check collection sizes |
+| StackOverflowError | Infinite recursion | Find recursive call with no base case |
+| Connection refused | Dependency service not running | Check config URLs, start dependent service |
+| Authentication/401 | Token expired, wrong config, missing header | Verify token, check auth config |
+| Permission denied/403 | Wrong role in request or misconfigured access rules | Check auth annotations and role config |
+| Database constraint violation | Duplicate key, missing FK, null in NOT NULL column | Check data and constraint definitions |
+| Serialization error | Circular reference, unmapped field | Check DTO/mapper configuration |
+| Test assertion mismatch | Wrong mock setup or wrong expected value | Re-read mock setup and assertion logic |
+
+### Test Failure Analysis
+
+```
+Expected: {expected value}
+Actual  : {actual value}
+  at {test class}.{method}(line {N})
+```
+
+1. What is the gap between Expected and Actual?
+2. Trace to the exact test line
+3. Is the mock setup correct? (`given(...).willReturn(...)`)
+4. Is the assertion logically correct?
+
+### Output
+
+```
+/debug Analysis
+
+## Error Encountered
+{description}
+
+## Root Cause
+{technical explanation}
+
+## Location in Code
+File: {path}
+Line: {line number if available}
+Layer: {Controller/Service/Repository/Test}
+
+## Suggested Fix
+{specific code change}
+
+## Related Rule
+See CLAUDE.md §{section} — {section name}
+
+## Next Step
+- To fully fix → run /fix-bug {TICKET_ID}
+- Just needed analysis → done here
+```
+
+{{include:steps/report-footer.md}}
+
+---
+
+## /validate-traces — Traceability Coverage Matrix
+
+Read-only check of coverage between specs, code, and tests.
+
+### Input
+
+Accept: domain name or specific UC-ID.
+
+### Process
+
+1. **Read .feature files** → collect all SC-IDs for the domain/UC
+2. **Scan controller files** → find `@trace.implements` tags
+3. **Scan test files** → find `@trace.verifies` tags
+4. **Read `.trace/{domain}/{prd-slug}/{UC-ID}.tsv`** → check for drift (scenario text changed since last code gen)
+
+### Output
+
+```
+/validate-traces — {domain}
+
+| UC-ID | SC | Spec | Code | Test | Status |
+|-------|----|------|------|------|--------|
+| {UC}-UC1 | SC1 | ✅ | ✅ {ControllerClass} | ✅ {TestClass} | synced |
+| {UC}-UC1 | SC2 | ✅ | ✅ {ControllerClass} | ❌ MISSING | gap |
+| {UC}-UC2 | SC1 | ✅ | ❌ MISSING | ❌ MISSING | gap |
+
+Gaps:
+  Missing code: [{UC-ID}-SC2, {UC-ID2}-SC1]
+  Missing tests: [{UC-ID}-SC1]
+  Drifted (spec changed, code not updated): [{list}]
+
+Recommendations:
+  - /generate-code {UC-ID} SC2
+  - /dev-gen-test {UC-ID}
+  - Re-run /generate-code {UC-ID} for drifted scenarios
+```
+
+{{include:steps/report-footer.md}}